Recently, ceramics composed principally of silicon nitride have found significant use as ceramic components for engine components or as vessel coatings. This material is known to have many good characteristics, such as high oxidation resistance at high temperatures (1400.degree. C.), good mechanical strength at high temperatures (1400.degree. C.), and good hardness at high temperatures.
Strength of this material is related to density and it has been found that the densification property of silicon nitride sintered under atmospheric pressure is very inferior. Therefore, it has been considered important to employ high pressure when a product of good strength is desired. This is routinely referred to as hot pressing of silicon nitride. However, in spite of the use of hot pressing, the bend strength of simple silicon nitride has not been as high as desired at high temperatures. Accordingly, other avenues of strength improvement have been sought, such as through the use of additives which operate as a low temperature liquid phase to facilitate densification. Although the Si.sub.3 N.sub.4 material became more dense, its resistance to thermal cracking and creep at high temperatures was lowered. A glassy phase in most instances was formed in the grain boundaries of the Si.sub.3 N.sub.4 matrix. The room temperature strength was increased, but at the expense of other physical properties of which the ceramicist was unaware.
These added materials have included relatively large amounts of chromium oxide, zinc oxide, nickel oxide, titanium oxide, cerium oxide, magnesium oxide, yttrium oxide and others, ranging in excess of 20% by weight of the matrix material. With these particular additives, the Si.sub.3 N.sub.4 tends to form a structure having a strength level which does not usually exceed 50 ksi at high temperatures. In one instance (U.S. Pat. No. 3,830,652 to Gazza) did the prior art obtain strength levels in excess of 50 ksi. In this instance, the concern was for physical characteristics useful for turbine elements: hardness, oxidation resistance (inertness) and transverse rupture strength. Gazza explored metal oxide additives to a Si.sub.3 N.sub.4 system which ranged in amounts related solely to machine element usage. The grain boundary phase was identified by Gazza as glassy in a companion article which appeared in the Journal of the American Ceramic Society, Vol. 56, No. 12, p. 662 (1973).
Commercial cutting tools today exhibit the same or better physical properties that were the focus of Gazza's work. For example, commercial Al.sub.2 O.sub.3 or TiC tools have excellent hardness at high temperatures, have high resistance to oxidation, and have transverse rupture strengths at high temperatures which range up to 100,000 psi. Tool designers, rightly or wrongly, have considered strength the most important feature because of the necessity to withstand the forces imposed on the tool material by the tool fixture and by the resistance of the stock material, particularly at heavy depths of cutting. These forces become unusually exaggerated when cutting ferrous material such as cast iron at high speeds and feeds. Without increased strength, it is believed by those skilled in the art that further improvements in tool life cannot be achieved. Since the strength level of Si.sub.3 N.sub.4 is equal to or lower than commercial materials now available, it has been rejected as a tool material candidate with little hope of improving tool life.
In only one known prior instance has the art attempted to employ Si.sub.3 N.sub.4 directly as a cutting tool material, and this was for use only on hypereutectic aluminum alloys. This attempt is set forth in Japanese Pat. No. 49-113803 to Ohgo (Oct. 30, 1974), appearing in Chemical Abstracts, Vol. 84, p. 286 (1976). In this work, silicon nitride was sintered (as opposed to hot pressing) and metal oxide spinels were employed in solid solution in the silicon nitride matrix. The spinels were formed by a mixture of divalent and trivalent metal oxides (including magnesium oxide and Y.sub.2 O.sub.3). However, the molar percentage of the spinel metal oxide in the material was taught to be 10-40%. The author experienced difficulty in obtaining good sintering density when the molar percentage fell below 10. The highest density achieved was 3.18 g/cm.sup.3. A two-step method was used by this Japanese author requiring first a heating of the metal oxide powders to 1300.degree.-1600.degree. C. for 3-10 hours to form the spinel. The spinel was pulverized and mixed with a silicon nitride powder which, in turn, was sintered to form cutting tools. Only a quarternary system was employed involving Si.sub.3 N.sub.4, SiO.sub.2, MgO and Y.sub.2 O.sub.3. This produced many secondary phases which weakened the physical characteristics, particularly strength, thermal conductivity, and increased the thermal coefficient of expansion. A loss of these physical characteristics would make it most difficult to obtain even equivalent performance to commercially available tools when applied to a rigorous cutting environment such as interrupted cutting on cast iron. The cutting operation of the Japanese patent was of very short duration (two minutes) of continuous machining, and at low metal removal rates (cutting speeds of 1000 sfm, 0.012 inches per rev. of feed, 0.060 inches of depth of cut, and metal removal of 8.64 in.sup.3 /min).
This type of test information, of course, did not investigate cutting applications where large forces are applied to the tool, did not investigate the elimination of spinel additives, did not investigate heavy cutting against rough surfaces such as cast iron, nor continuous cutting for periods of several hours or greater, not did it explore intermittent, interrupted high speed cutting at speeds of 4000-5000 sfm at heavy feeds and depths of cutting. The demonstrated wear of 0.006-0.008 inches (in Ohgo's work) for two minutes of cutting time is highly excessive when compared to the goals of the present invention. Therefore, this work did not demonstrate that Si.sub.3 N.sub.4 possessed sufficient characteristics to be used as a tool material on ferrous materials which apply large bend forces to the tool.
Moreover, the art has been possessed of sufficient knowledge in the making of Si.sub.3 N.sub.4 with additives for many years. During this long term, no effort was made to apply this material as a cutting tool against cast iron. This tends to support the contention of this invention that if tool life is dramatically increased for certain Si.sub.3 N.sub.4 composites when used for machining cast iron, there must be some unobvious characteristics independent of strength that layed undiscovered to promote this new use.